Chronic nonvalvular Atrial fibrillation (AF) is a very common and costly public health problem. AF may reduce the quality and duration of life, both, particularly by increasing the risk of devastating thromboembolic strokes. Anticoagulants, including vitamin K antagonists (VKAs, e.g., warfarin) and non-VKA anticoagulants (NOACs, e.g., dabigatran, -xabans), are very effective, but underused treatments. Anticoagulation underuse results form challenges clinicians face in prescribing these drugs and patients face in implementing their use in their lives. Recognizing these challenges, leading cardiovascular societies in 2014 formulated a class I recommendation of shared decision making (SDM) for the individualization of anticoagulation therapy in at-risk patients with AF. However, there are no proven interventions to support the implementation of this recommendation, and no evidence of the effect of SDM on the problem of initiating and implementing anticoagulation. With over a decade of experience, we have demonstrated the practical impact of SDM interventions in other contexts. Building on this experience, we propose to use an SDM tool we developed using user-centered design, ANTICOAGULATION CHOICE, to promote SDM about anticoagulation between patients with AF and their clinicians. Using a randomized trial design, we plan to determine the impact of using this tool on the quality of SDM and on the rate of anticoagulation and patient adherence to this therapy. By generating reliable practice-based evidence, this trial can provide feedback to the guideline panel about the impact of SDM on the quality of individualized anticoagulation care. We propose to enroll 999 adult patients with chronic nonvalvular AF deemed at high risk of thromboembolic strokes (CHA2DS2-VASc score ? 1, or 2 in women) and receiving care in academic, community, or safety net clinics. We plan to randomly assign them to care with or without a new SDM tool, ANTICOAGULATION CHOICE. Trial outcomes include the quality of SDM (primary endpoint, aim 1), the rate of anticoagulation, the distribution of anticoagulants chosen, and patient adherence to anticoagulation at 12 months (aim 2). If successful, this work will reduce the underuse of anticoagulation therapy and improve the care and outcomes of millions of patients with AF.